Search

Measurements of surface thermal infrared (TIR) radiance that are made to extract temperatures display strong directional anisotropy effects. Directional brightness temperature (BT) models that describe this anisotropic behavior of TIR emissions can be applied to separate component temperatures using multi-angle observations. The surface temperature differences that occur between sunlit and shaded areas and the leaf clumping phenomenon jointly affect the directional signatures of out-of-canopy directional BTs. However, these factors are not fully considered in existing directional BT models. This paper therefore extends the FR97 analytical model to 1) a four-component scene containing sunlit and shaded soil and leaves by incorporating the effective emissivity values of the sunlit and shaded parts and 2) row-planted crop and forest canopies by introducing a leaf clumping index. The proposed model was assessed using a synthetic dataset that was generated by the Thermal Radiosity-Graphics Combined Model (TRGM) under various conditions. The evaluation results indicated that the proposed model performed as well as the Scattering by Arbitrarily Inclined Leaves (4SAIL) model over continuous canopies with root mean squared errors (RMSEs) lower than 0.3 °C. Over non-continuous crops and forests, the behavior of the proposed model displayed improved agreement with the TRGM with RMSEs lower than 0.65 °C. The proposed model also displayed a robust performance over both the maize and pine canopies, which was evaluated against the directional anisotropy of measured datasets that were collected at the Huailai remote sensing test site and the Institut National de la Recherche Agronomique (INRA), respectively. From these points, the proposed model has potential for component temperature inversion and rapid assessment of TIR angular effects. [ABSTRACT FROM AUTHOR]

Thermal radiation directionality (TRD) characterizes the anisotropic signature of most surface targets in the thermal infrared domain. It causes significant uncertainties in estimating surface upward longwave radiation (SULR) from space observations. In this regard, kernel-driven models (KDMs) are suitable to remove TRD effects from remote sensing dataset as they are computationally efficient. However, KDMs requires simultaneous multi-angle observations as inputs to be well calibrated, which yields a difficulty with geostationary satellites as they can only provide a single-angle observation. To overcome this issue, we proposed a six-parameter time-evolving KDM that combines a four-parameter SULR diurnal variation model and a two-parameter TRD amplitude model to correct the TRD effect for single-angle estimated SULR dataset of geostationary satellites. The significant daytime TRD effect when solar zenith angle is within 60° can be effectively eliminated. The modeling accuracy of the time-evolving KDM is evaluated using a simulated SULR dataset generated by the 3D Discrete Anisotropic Radiative Transfer (DART) model; the TRD correction method based on the new time-evolving KDM is validated using a two-year single-angle estimated SULR dataset derived from data of the Advanced Baseline Imager (ABI) onboard Geostationary Operational Environmental Satellite-16 (GOES-16) against in situ measurements at 20 AmeriFlux sites. Results show that the proposed time-evolving KDM has a high accuracy with an R2 > 0.999 and a small RMSE = 1.5 W/m2; the TRD correction method based on the time-evolving KDM can greatly reduce the GOES-16 SULR uncertainty caused by the TRD effect with an RMSE decrease of 4.5 W/m2 (22.1%) and mean bias error decrease of 7.9 W/m2 (62.7%). Hence, the proposed TRD correction method is practically efficient for the operational TRD correction of SULR products generated from the geostationary satellites (e.g., GOES-16, FY-4A, Himawari-8, MSG). • A 6-parameter time-evolving kernel-driven model for SULR was proposed. • The multi-temporal observations of ABI/GOES-16 are utilized to drive new model. • The TRD effect of SULR product of geostationary satellite was corrected for the first time. • The RMSE and MBE decreased by 22.1% and 62.7% for the ABI/GOES-16 SULR. [ABSTRACT FROM AUTHOR]

Striping noise is a well-known phenomenon that arises in most multi-detector optical imaging instruments. Such noise affects both visual interpretation and quantitative analysis. Therefore, destriping is an essential step before absolute calibration and image interpretation. Histogram matching is one of the most popular algorithms used to reduce striping. The assumption underlying histogram matching is that each detector has the same gray level distribution. This assumption is easily satisfied when the image is sufficiently large, but it often cannot be satisfied for small images. An improved histogram matching algorithm based on sliding windows is proposed in this paper. The algorithm presupposes that the gray level distribution of each column (taking the vertical striping noise as an example) is similar to the gray level distribution of the column-centered local area. The size of the local area is determined by a histogram growing algorithm. Compact High Resolution Imaging Spectrometer (CHRIS), Moderate Resolution Imaging Spectroradiometer (MODIS) and Hyperspectral Imager (HSI) images were used to test the new and traditional algorithms. These destriping results were compared using improvement factors, inverse coefficients of variation and mean profiles. The results of the comparison indicate that the improved histogram matching algorithm has obvious advantages over traditional method. [ABSTRACT FROM AUTHOR]

This paper proposes a new welding method for joining non-ferrous metals: resistance heat assisted ultrasonic welding. Resistance heat generated by the electric Joule effect is used as an additional electrical energy source to assist ultrasonic welding process. A comparison is conducted between two dissimilar Al–Cu joints, one produced by ultrasonic welding and the other by resistance heat assisted ultrasonic welding. In the resistance heat assisted ultrasonic process, the peak power of ultrasonic vibration increases significantly. The interfacial reaction between aluminum and copper is studied as a function of the current. The thickness of the intermetallic compound layer, which is predominantly composed of CuAl 2 , increases with the current. At a relatively high current (1500 A), resistance heat assisted ultrasonic welding produced a dendritic solidification microstructure at the interface, due to the occurrence of a eutectic reaction, α -Al + θ → L , during the welding process. The influence of current on the mechanical properties of the joints is also discussed. [ABSTRACT FROM AUTHOR]

Radiometric measurements in the Thermal Infrared (TIR) domain exhibit an angular variation over most surface types, known as the Thermal Radiation Directionality (TRD) phenomenon. A primary objective of the ongoing development of TRD physical models is to perform a correction of the angular effects to obtain comparable land surface temperature products. In practice, it is advised to handle only the models having a limited number of input parameters for the purpose of operational applications. The use of semi-empirical kernel-driven models (KDMs) appears to be a good tradeoff between physical accuracy and computational efficiency as it was already demonstrated through a broad usage in the optical domain. It remains that the existing state-of-the-art 3-parameter TIR KDMs (RossThick-LiSparseR, LiStrahlerFriedl-LiDenseR, Vinnikov, and RoujeanLagouarde) underestimate the hotspot phenomenon, especially for continuous canopies marked by a narrow peak. In this study, a new general framework of TIR kernel-driven modeling is proposed to overcome such issue. It is a linear combination of three kernels (including a base shape kernel, a hotspot kernel with adjustable width and an isotropic kernel) with the ability to simulate the bowl, dome and bell shapes in the solar principal plane. Four specific 4-parameter models (Vinnikov-RoujeanLagouarde, LiStrahlerFriedl-RoujeanLagouarde, Vinnikov-Chen, and LiStrahlerFriedl-Chen, named "base shape kernel - hotspot kernel") within the new framework were studied to assess their abilities to mimic the patterns of the directional brightness temperature for both continuous and discrete vegetation canopies. These four 4-parameter KDMs and four 3-parameter KDMs were comprehensively evaluated with 306 groups of simulated multi-angle datasets generated by a modernized analytical 4-stream radiative transfer model based on the Scattering by Arbitrarily Inclined Leaves (4SAIL), and a Discrete Anisotropic Radiative Transfer (DART) model considering different solar zenith angles (SZA), canopy architectures and component temperatures, and 2 groups of airborne measured multi-angle datasets over continuous maize and discrete pine forest. Results show that the four 4-parameter KDMs behave better than the four existing 3-parameter KDMs over continuous canopies (e.g. R2 increases from 0.661~0.970 to 0.940~0.997 and RMSE decreases from 0.17~0.71 to 0.07~0.16 when SZA = 30°) and discrete canopies (e.g. R2 increases from 0.791~0.989 to 0.976~0.996 and RMSE decreases from 0.10~0.84 to 0.08~0.21 when SZA = 30°). The new general framework with four parameters (three kernel coefficients and an adjustable hotspot width) improves the fitting ability significantly, compared to the four existing three-parameter KDMs, given the addition of one more degree of freedom. Results show that the coefficients of the base shape kernel, hotspot kernel and isotropic kernel are related to the temperature difference between leaf and background, temperature difference between sunlit component and shaded component, and the nadir brightness temperature, respectively. However, the estimated hotspot width depends on vegetation structure. The new kernel-driven modeling framework has the potential to be a tool for angular correction of multi-angle satellite observations and angular optimization of future multi-angle TIR sensors. • A general framework of thermal infrared kernel-driven modeling is proposed. • It contains an isotropic kernel, a K BaseShape , and a K Hotspot with adjustable width. • Three new 4-parameter models (LSF-RL, Vinnikov-Chen, and LSF-Chen) are developed. • The four 4-parameter models can simulate the bowl, dome and bell patterns accurately. • The four 4-parameter models behave much better than existing four 3-parameter models. [ABSTRACT FROM AUTHOR]

The Earth surface thermal infrared (TIR) radiation shows conspicuously an anisotropic behavior just like the bi-directional reflectance of visible and near infrared spectral domains. The importance of thermal radiation directionality (TRD) is being more and more widely recognized in the applications because of the magnitude of the effects generated. The effects of TRD were originally evidenced through experiments in 1962, showing that two sensors simultaneously measuring temperature of the same scene may get significantly different values when the viewing geometry is different. Such effect limits inter-comparison of measurement datasets and land surface temperature (LST) products acquired at different view angles, while raising the question of measurement reliability when used to characterize land surface processes. These early experiments fostered the development of modeling approaches to quantify TRD with the aim of developing a correction for Earth surface TIR radiation. Initiatives for pushing the analysis of TIR data through modeling have been lasted since 1970s. They were initially aimed at mimicking the observed TIR radiance with consideration of canopy structure, component emissivities and temperatures, and Earth surface energy exchange processes. Presently, observing the Earth surface TRD effect is still a challenging task because the TIR status changes rapidly. Firstly, a brief theoretical background and the basic radiative transfer equation are presented. Then, this paper reviews the historical development and current status of observing TRD in the laboratory, in-situ, from airborne and space-borne platforms. Accordingly, the TRD model development, including radiative transfer models, geometric models, hybrid models, 3D models, and parametric models are reviewed for surfaces of water, ice and sea, snow, barren lands, vegetation and urban landscapes, respectively. Next, we introduce three potential applications, including normalizing the LST products, estimating the hemispheric upward longwave radiation using multi-angular TIR observations and separating surface component temperatures. Finally, we give hints and directions for future research work. The last section summarizes the study and stresses three main conclusions. • A review of thermal radiation directionality models in past 50 years is performed. • Current status of multi-scale multi-angle observation is reviewed. • Fourteen quantities for describing the TRD effect are summarized. • Three potential applications and five future directions are suggested. [ABSTRACT FROM AUTHOR]

The land surface emissivity (LSE) in the MYD21 product contains the effects of viewing zenith angle. The influence of the angular variation of LSE on the surface upwelling longwave radiation (SULR) estimation is still unexplored at the satellite scale. We performed statistical analyses of MYD21 emissivity retrievals over different land surface types for three longwave bands centred around 8.55 μm (Band 29), 11 μm (Band 31) and 12 μm (Band 32), respectively. A look-up table was generated to describe the angular variations for both single-band and broadband emissivities. The results showed that the angular variation of directional emissivity in Band 29 could reach up to 0.03, but was <0.01 for Bands 31 and 32. The angular variation in broadband emissivity was intermediate to that for individual bands. In all cases, the directional emissivity was greatest and symmetric around nadir. By integrating the directional broadband emissivity, the influence of angular variation of the LSE on estimated SULR was quantified using simulation and measurements at seven stations from the US surface radiation budget network (SURFRAD). The difference between the directional and integrated hemispheric broadband emissivity was within 0.01. As a result, the influence of angular variation of LSE on the SULR estimation was modest. For the SURFRAD stations, the differences of root-mean-square error (RMSE) before and after considering the angular variation of LSE were generally <1 W m−2. We conclude that the angular variation of broadband emissivity is not pronounced because of the small linear weight for Band 29 in the calculation of broadband emissivity. Ignoring the anisotropy of emissivity does not introduce large errors in SULR estimation generally. • MODIS 8.5–12 μm emissivities retrieved using a physical algorithm were analysed. • Anisotropy in emissivity was demonstrated and quantified. • Emissivity differences were found between surface types, day/night, and seasons. • Ignoring emissivity anisotropy does not affect longwave radiation estimation greatly. [ABSTRACT FROM AUTHOR]

Surface upwelling longwave radiation (SULR) is a major component of the Earth's radiation budget and directly influences the retrieval of evapotranspiration (ET) in the terrestrial ecosystems. Land surface temperature (LST) is an Essential Climate Variable (ECV) for direct estimation of SULR. However, accurate retrieval of SULR from satellite observations may be severely hindered by the anisotropic properties of land surface targets since most of them show marked angular variations in LST. This study aims at investigating the magnitude and impact factors of the directional effects of LST on SULR estimation over vegetated surfaces given that angular variation in emissivity has a limited impact on SULR estimation over most land surface types. It follows an attempt to correct for such effects in SULR estimation. We further explore the possibility to find a viewing direction at which SULR estimated from the directional LST can surrogate the hemispherical integration. To do so, a parametric model mimicking LST anisotropy with the hot spot is incorporated into the physical temperature-emissivity method. Two widely used Moderate Resolution Imaging Spectroradiometer (MODIS) LST products (i.e., MYD11_L2 and MYD21_L2) are analyzed. SULR estimates before and after correcting for LST directionality are compared with in-situ measurements acquired at 15 sites from the FLUXNET and SURFRAD networks in different regions. Our analysis reveals that LST directional effects on SULR estimation exhibit diurnal and seasonal variations, which are substantial in spring and summer for the daytime. The effects are negligible (<5 W m−2) in autumn and winter for the daytime except for in arid and semiarid regions. For the night-time, the effects are insignificant over all the biomes. Using MYD21 LST, after correction, the average root-mean-square error (RMSE) and bias of SULR estimates for all sites decrease by 8 and 8.34 W m−2 in spring, and by 8.9 and 12.13 W m−2 in summer. Using MYD11 LST, after correction, the average RMSE is between 10 and 15 W m−2 and the average bias is close to zero in all seasons. The RMSE and absolute bias of SULR estimates for sites with low to moderate vegetation (LAI <3) is lowered substantially (7–14 W m−2) after correction. Interestingly, SULR estimates from LST viewed at 54° backward and hemispherically integrated are close, with differences <3 W m−2 at most of the sites. These findings support a strategy for SULR estimation in ET retrieval over vegetated surfaces from directional LST. • LST directional effects on SULR are substantial in spring and summer for daytime. • SULR accuracy after considering LST directionality is markedly improved. • The improvement is the greatest for sites with low to moderate vegetation. • SULR from LST viewed at 54° backward and hemispherically integrated are close. [ABSTRACT FROM AUTHOR]

Land surface temperature (LST) is an essential climate variable (ECV) which can be estimated from appropriate measurements of the surface thermal infrared (TIR) radiance. LST varies on a very short time scale and closely depends on the illumination and scan angles considered. To fully exploit LST products, a method for reconstructing the temporal profile and the angular dependence at the same time is proposed here. A combined visible-thermal envelope method (VT-KDTC) is built using kernel-driven (KD) and diurnal temperature cycle (DTC) models, referring to the surface structure and thermal factors, respectively. To demonstrate the reliability of the approach, TIR data from the geostationary satellite Himawari 8 are combined with visible and near-infrared (VNIR) data from the polar orbit satellite Sentinel-3A/3B. In addition to satellite observations, a synthetic dataset from the Soil Canopy Observation, Photochemistry and Energy Fluxes (SCOPE) model is also generated. Considering an anisotropy model in addition to the DTC model leads to a method displaying a better ability to simulate LSTs with a root mean squared error (RMSE) of 0.48 K against the original satellite results, compared to only the DTC model up to 1.44 K. By utilizing the field measurements as a reference, the reconstructed results are improved with a total bias of 0.72K and an RMSE of 2.58 K. Compared to the original results without correction, approximately 41% and 10% decreases are obtained in bias and RMSE, respectively. Our proposed method can also achieve LST downscaling supported by the higher spatial resolution of VNIR data when the temperature difference is assumed to be homogeneous within the coarse pixels. Thus, a simple achievable solution can be used for temperature reconstruction to enhance the quality of the LST product. • Reconstruct the angular and temporal change of LST. • Apply the VNIR and TIR observations together. • Combine kernel driven and diurnal temperature cycle models. • The spatial resolution of LST can be further improved. [ABSTRACT FROM AUTHOR]

Land surface temperature (LST) is a vital parameter for the achievement of the surface energy budget and in thorough investigations of water cycle processes. Lightweight thermal infrared (TIR) sensors onboard unmanned aerial vehicles (UAVs) are rapidly becoming key instruments for extracting high-resolution LSTs given the flexibility they offer in capturing different scales. With this expansion, there has been increasing concern regarding the growing demand to obtain a mapping of normalized LST given the directional anisotropy (DA) of surface fine-scale emissions. To date, this topic suffers from a lack of deep analysis and practical solutions for characterizing the DA of fine-scale TIR data from UAV measurements over tree and crop canopies. In this paper, the first objective was to understand the pattern of brightness temperatures (BTs) DAs at a high spatial resolution by using UAV-based multiangle observations and three-dimensional (3D) radiative transfer model simulations. This study highlighted the need for first performing an angular normalization of the BTs of fine-scale pixels prior to any application, as these were easily affected by adjacent pixels and displayed broad spatial variability from 0.5 °C to 5.0 °C due to 3D occlusion. The second objective of the present study was to appraise the reliability of a modified kernel-driven model, in comparison to the model from which it was derived, with an additional kernel designed to mimic the adjacency effect, plus, a quadratic function used to simplify the estimate of the directional emissivity kernel. The root mean square error of the best fit between the measured UAV dataset and the modified kernel-driven model was approximately 0.65 °C, which proves its efficiency since the DA indexes of the BTs were about 1.40 °C. This outlined the role of the model to normalize from directional effects the camera image pixels and thereby deliver fine-scale BTs. In addition, results from LESS simulations also demonstrated the good performance of the modified kernel-driven model for simulating the DAs of thermal emissions for both tree and row-planted scenes. Index Terms—Land surface temperature, UAV, directional anisotropy, high spatial resolution. • Highlight the understanding of directional anisotropies for fine-scale TIR data. • Propose a new kernel for 3D occlusion between adjacent pixels. • Propose a simplified version for a directional emissivity kernel. • Explore a potential angular normalization strategy for UAV TIR data. [ABSTRACT FROM AUTHOR]

The enhancement of the living conditions in big cities since the end of the last century is closely related to changes in the thermal environment and besides in urban microclimate, particularly for metropolitan areas. In this context, a knowledge of the spatial and temporal variability of urban heat island (UHI) became an increasing matter of concern, which can be measured from land surface temperature (LST). Actually, LST can be derived from thermal infrared (TIR) remote sensing observations to ensure the necessary spatial and time frequency coverage. But a full exploitation of satellite TIR data cannot be achieved without accounting for the strong anisotropy of urban landscape. Hitherto, a poor investigation was focused on the modeling and the analysis of the directional anisotropies of LSTs considering the complexity of urban building surfaces (e.g., heterogeneity of building morphology and temperature distribution) whereas it is fundamental to establish reliable critical indicators derived from energy balance. Herein, we propose an analytical model to simulate the angular signatures of urban temperatures, in which the geometric optical theory is considered to model the direct radiances of the main components (i.e., sunlit and shaded street, roof and wall). The built model assumes a random distribution of low/middle-rise and high-rise buildings, which depicts realistically the heterogeneity of urban architectural distribution. We evaluated the proposed model using both measured datasets from airborne and satellite sensors and a simulated dataset from a 3D ray-tracing model so-called discrete anisotropic radiative transfer (DART). Results indicate that 1) the proposed model is effective for simulating directional anisotropies of LSTs, with a root mean square error (RMSE) lower than 0.90 ° C and R 2 > 0.49 for comparison with measured datasets; and 2) the directional anisotropies of LSTs are significantly affected by variations in building height, with values possibly exceeding 1.5 ° C. The proposed model can be perceived as a useful tool to analyze the contribution of each component and to assess the impact of urban structure. Furthermore, it can serve to improve urban radiation budget estimations in mixed pixels. • Analyze the directional anisotropy of urban temperature. • Consider the building shape and multiple scattering effect. • Propose an analytical model to simulate angular temperatures. • Quantify the impact of urban heterogeneity on temperature. [ABSTRACT FROM AUTHOR]

Characterizing drought is mandatory for a thorough monitoring of crop growth. Such information can be inferred from radiometric measurements in optical and thermal spectral ranges. The Temperature Vegetation Dryness Index (TVDI) presents the advantage to mix spectral information since it associates land surface temperature (LST) with normalized difference vegetation index (NDVI). NDVI is known to be weakly impact by directional effects, which is not the case of TVDI as the accuracy assessment of LST can be severely hampered directional anisotropy (DA). In this paper, DA feature of TVDI is analyzed using the kernel-driven model Vinnikov-Li (VinLi) that serves to perform through simulations and remove DA. The study uses both airborne and satellite datasets. TVDI was found to be angularly dependent, with a possible uncertainty larger than 15%. Compared against TVDI without angular correction, the normalized results displayed a better correlation with soil moisture content, and the root mean squared error (RMSE) of estimated soil moisture content decreased obviously, from 0.050 m 3/ m 3 to 0.045 m 3/ m 3 for airborne data over maize canopies, from approximately 0.036 m 3/ m 3 to 0.030 m 3/ m 3 for satellite data over cropland canopies, and from approximately 0.032 m 3/ m 3 to 0.028 m 3/ m 3 for satellite data over grassland canopies. The evaluation was carried out using the complied synthetic datasets based on the Soil Canopy Observation, Photochemistry and Energy Fluxes (SCOPE) model. These also confirmed the occurrence of significant DAs of TVDIs and good performance of VinLi in reducing DA of TVDI by >50%. Based on the VinLi method, an angular-independent TVDI appears necessary to reduce uncertainties originated from illuminating and viewing geometries for the purpose of agriculture monitoring with possible drought episodes. Although further evaluation is still needed using surface measurements, the outcomes support a confident use of TVDI for observations from satellite and airborne/unmanned aerial vehicle. • Analyze the directional anisotropy of TVDI. • Propose a parametric method to simulating angular TVDIs. • Extend the kernel-driven strategy to TVDI. • Introduce an angular-independent TVDI. [ABSTRACT FROM AUTHOR]

Abstract: In this study, the Visible Infrared Imager Radiometer Suite (VIIRS) land surface temperature (LST) environmental data record (EDR) and Moderate Resolution Imaging Spectroradiometer (MODIS) L2 swath LST products (collection 5) from both the Terra and Aqua satellites were evaluated against ground observations in an arid area of northwest China during the Heihe Watershed Allied Telemetry Experimental Research (HiWATER) experiment. Four barren surface sites were chosen for the evaluation, which took place from June 2012 to April 2013. The results show that the current VIIRS LST products demonstrate a reasonable accuracy, with an average bias of 0.36K and −0.58K and an average root mean square error (RMSE) of 2.74K and 1.48K for the four sites during daytime and nighttime, respectively. The accuracy of the nighttime LST is much better than that of daytime. Furthermore, it was also found that the VIIRS split-window (SW) algorithm provides better performance than the VIIRS dual split-window (DSW) algorithm during both daytime and nighttime. For MODIS LST products, the results show that both Terra and Aqua MODIS C5 LST products underestimate the LST for the four barren surface sites at daytime, and the biases and RMSEs are much larger for Aqua, with biases varies from −0.91K to −3.13K for Terra and from −1.31K to −3.76K for Aqua. [Copyright &y& Elsevier]

Land surface temperature (LST) is listed as an essential climate variable (ECV) and supports quantitative estimates of the energy budget while serving as a proxy for measuring the effects of climate change and extreme events. Forested areas are considered a major land unit impacted by temperature rise; therefore, thorough monitoring is mandatory. An accuracy assessment of the LST of forests must consider their directional anisotropy (DA). This latter can be well depicted by thermal infrared (TIR) radiative transfer models, but the problem is complex for forests because many of the shaded areas generate multiscale gradients of temperature. In this paper, we adapted a mature and widely used visible and near-infrared (VNIR) radiative transfer model called forest reflectance and transmittance (FRT) to enhance the characterization of the DA of forest temperature. In the FRT model, the vertical heterogeneity of the forest is quantified by using the discrete elements of multilayer scene components (i.e., the tree crown, trunk, understory vegetation, and soil), thus inferring vertical thermal gradients. The Planck function and spectral-invariant theory are considered to assess the thermal emissions of the scene components and their multiple scattering processes. The FRT model is validated using directional forest brightness temperatures (BT) measured from an unmanned aerial vehicle (UAV) and simulated by using the three-dimensional ray-tracing LESS (large-scale remote sensing data and image simulation framework over heterogeneous 3D scenes) model. The results show that FRT behaves reliably since the root mean square error (RMSE) is lower than 1.0 °C for UAV measurements obtained at 09:20 and 13:10 and with coefficients of determination ( R 2 ) larger than 0.74 and 0.56, respectively; these results are better than the simulated results by existing models. Moreover, the comparison with ray-tracing simulations was also deemed satisfactory. According to the analysis, large variations in BT DAs may appear for different forests and seasonal changes staged by structural and thermal stratification, thus indicating the necessity of using the FRT model for complex and dynamic forest canopies. • Extend the VNIR FRT model to TIR applications. • Consider forest structural and thermal stratification • Analyze BTs DAs with different forests and seasonal changes. • Introduce a spectral-invariant theory to FRT. [ABSTRACT FROM AUTHOR]